Every great AI output starts with a great input. CRISP gives your prompts structure so AI stops guessing and starts delivering.
Background situation. What's happening, what market, what deal stage.
"I'm preparing for a meeting with a system integrator in Thailand who currently uses Cambium..."
Who should the AI act as? This shapes the tone and expertise level.
"Act as a senior wireless infrastructure analyst with 15 years in telecom..."
What do you want to achieve? Be specific about the goal.
"Help me prepare 5 discovery questions that uncover real pain points..."
Constraints, format, length, number of items, audience.
"Format as a one-page brief. Use bullet points. Executive-level language..."
Your product context. What you sell, key differentiators, verticals. This is where your knowledge files do the heavy lifting — upload product specs, competitive data, and case studies so every prompt starts with Radwin context built in.
Set this once in ChatGPT → Settings → Personalization → Custom Instructions. Every conversation starts with your context automatically.
A Custom GPT is a reusable AI tool with persistent instructions and knowledge files. Pick the one that matches your biggest pain point and build it in 3 steps.
Your market intelligence analyst. Researches markets, identifies key players, system integrators, and opportunities before business trips.
Validates distributors and system integrators before you invest time. Researches their track record, financial stability, and strategic fit.
Handles objections, builds competitive positioning, and prepares you for tough conversations.
Generates battle cards and competitive comparisons on demand. Keeps your competitive intelligence fresh.
Finds and qualifies prospects. Turns vague territory plans into actionable target lists.
The difference between a junior AI user and a power user: what you feed it. Create two files and upload them to your Custom GPTs. Every time you add information, your AI gets sharper. Your colleague who doesn't do this stays at zero.
Everything AI needs to know about Radwin to give you accurate, contextual answers.
Include:
• Product lines: FIM, JET, 2000 series, MultiHaul — key specs and use cases
• Target verticals and typical project sizes
• Competitive advantages (vs Cambium, Ubiquiti, Siklu, fiber)
• Pricing positioning (premium, value, TCO arguments)
• Recent wins and case studies
• Common objections and how you handle them
Update trigger: After every major win, product update, or new case study
A living document that makes your competitive intelligence compound over time.
Include:
• Competitor product lines and recent launches
• Known pricing (even estimates or ranges)
• Strengths and weaknesses per vertical
• Regions where they're strong/weak
• Common objections prospects raise about Radwin vs them
• Key deals lost or won against them (anonymized if needed)
Update trigger: After every competitive deal, lost deal debrief, or market news
Copy these into ChatGPT, replace the [brackets] with your real context, and get immediate value.
Google NotebookLM is your research assistant for your own documents. Unlike ChatGPT, it only answers based on what you upload — grounded, source-cited, no hallucinations.
• Analyzing RFPs, technical documents, or competitor spec sheets
• Researching across multiple uploaded market reports
• Getting source-cited answers (with page references)
• Comparing documents side by side
• When accuracy matters more than creativity
• Generating content (emails, proposals, presentations)
• Brainstorming and ideation
• Live web research (current market data)
• Building Custom GPTs with persistent context
• When speed and flexibility matter more than source accuracy